On Jan 22, 2008 11:01 PM, Peter Saint-Andre <[EMAIL PROTECTED]> wrote: > > Peter Saint-Andre wrote: > > Tobias Markmann wrote: > >> On Jan 22, 2008 10:14 AM, Richard Dobson <[EMAIL PROTECTED]> wrote: > >>> Kevin Smith wrote: > >>>> On Jan 22, 2008 6:02 AM, Tobias Markmann <[EMAIL PROTECTED]> > >>>> wrote: > >>>> > >>>>>> URL: http://www.xmpp.org/extensions/inbox/clientinfo.html > >>>>>> > >>>>> What are the enhancements of this XEP compared to XEP-0092? Why should > >>>>> one implement this XEP and not XEP-0092? Since both XEPs seem to do > >>>>> the same job I think there is missing a passage which is when to user > >>>>> or even if XEP-0092 is even to use in future. > >>>>> > >>>> The aim was to wrap this up inside caps hashes, so you wouldn't a) > >>>> need to query frequently, or b) pollute the presence packets with more > >>>> info, which was what other proposals did. > >>>> > >>> Plus even if you wern't implementing caps but were implementing disco > >>> you would get the version information for "free" without having to > >>> separately query for it. > >>> > >>> Richard > >>> > >>> > >>> > >> > >> Well, then the XEP should say something about that or even better that > >> this protocol should be used instead of XEP-0092 since it's generally > >> bad to have two standards with the same or nearly the same purpose. In > >> the end the target of a standardization organization is to have just > >> one protocol for a certain purpose otherwise there is implementation > >> overhead to remain interoperability. > > > > Yes, I'll add some text about that. I wrote it in a hurry last night as > > a follow-up to the XEP-0115 discussion. > > I added the following sentence: > > "This protocol is intended to replace Software Version [3] for > client-to-client information discovery (thus reducing or eliminating the > need for distinct software version requests) and also provides a format > that can be encapsulated into Entity Capabilities [4] notifications." > > http://www.xmpp.org/extensions/inbox/clientinfo.html > > > Peter > > -- > Peter Saint-Andre > https://stpeter.im/ > >
Yeah, that's exactly that what I thought was missing. Tobias
